Bundesgerichtshof (Sixth Civil Senate) 9 May 1995, NJW 1995, 2162, with case note.
This case is first published in the German Law Archive courtesy of:
Translated German Cases and Materials under the direction of Professors P. Schlechtriem, B. Markesinis and S. Lorenz
Translated by Mr Raymond Youngs, Southampton Institute
Facts: The defendant bottled and distributed carbonated mineral water in returnable glass bottles, either “standard spring-water bottles” or bottles of thicker glass and different shape. On 27 June 1990 the plaintiff, then nine years old, was fetching two bottles of the latter type from the cellar of her parents’ house. She had placed them on the floor outside the cellar in order to close the door and was about to pick them up when one of the bottles exploded. Splinters of glass entered her left eye and caused serious injuries which despite an operation reduced her sight to 60% and left her with astigmatism.
When bottles were to be refilled in the defendant’s factory the following process took place. Cases of empties were put on rollers and carried to a conveyor belt, where the bottles, still in the cases, were visually inspected by two of the defendant’s operatives, whose job it was to discard any intrusive or damaged bottles. Then a grab armed with rubber bulbs picked up the bottles, three at a time, and transported them to the washing point, where they were repeatedly sprayed with water. On leaving the washing point, still on the conveyor belt, they were inspected again visually by another employee. They then passed through the bottle-inspection unit, an electronically-operated machine which passed a beam of light through the base and mouth of each bottle, and if this disclosed any damage in those parts, the bottle was discarded. There was then a further visual inspection before the conveyor belt carried the bottles to the filling station. Before entering the filling station they were subjected to yet another visual inspection, and then entered the pressure chamber, where they were subjected to a pressure of 5 bar, a pressure one-third greater than would be exerted by the contents of the bottle once filled. On leaving the pressure chamber the bottles were filled, visually checked once again, and then labelled. The conveyor belt then carried them to the packing station where rubber grips lifted them and placed them in cases for consignment. About 15,000 bottles per hour were processed in this manner.
The plaintiff, relying on the report of the national materials laboratory in D. which he put in evidence, maintained that there was a fault in the glass at the site of the fracture – a chip about 4 mm. broad, which may well have existed at the time the bottle was delivered.
For her pain and suffering the plaintiff claimed an appropriate capital sum plus monthly instalments of DM 500. She also claimed a declaration that the defendant was bound to make good to her any material loss she might suffer as a result of the accident on 27 May 1990, except in so far as her claims may have vested in third parties.
Both lower courts dismissed her claim. Her appeal was now allowed.Reasons: The court below, whose decision is published in VersR 1995, 103, accepted the evidence that explosions of glass bottles filled with carbonated water are always due to damage to the surface of the bottle, and that spontaneous explosions occur through the spread of a very fine hairline crack. The very slightest physical contact, even, under certain circumstances, that of a warm hand, may be enough to cause the bottle to break. The court below held that in this case the bottle had such a hairline crack in it and that the defendant should have withdrawn it during the production process, but that it was at fault in failing to do so and letting the bottle get into circulation. Accordingly the precondition of a claim for damages for pain and suffering, namely that the defendant have been culpably in breach of a duty of care towards the public, was not satisfied. The explosive bottle was one which unavoidably “got away” despite the exercise of all appropriate precautions. Certainly before a manufacturer puts into circulation a product which entails particular risks he must take every care to make sure that it has no defect. But the defendant’s techniques of quality control were, according to the experts, up to the state of the art, and no system of control could absolutely ensure that no bottle with a hairline crack left the factory. According to the experts there was an irreducible residual risk in refilling glass bottles, and this case was an instance of it.
The court below was also of the view that the defendant was not liable for the future material loss of the plaintiff under the Product Liability Law of 15 Dec. 1989 (BGBl. I, 2198), since it followed from the reasons given for dismissing the claim in tort that in the current state of scientific and technical knowledge the defect in the bottle could not have been detected (II no. 5). The appellant’s criticisms of the judgment below are justified.
1. The court below was wrong to dismiss the plaintiff’s claim for damages for her material harm under 1 of the Product Liability Law.
a) The plaintiff rightly objects on procedural grounds to the finding that there was a hairline crack in the bottle, which exploded in the plaintiffs’ hand. There was no basis for such a finding. The plaintiff’s case was that there was a 4 mm. chip off the surface of the bottle, and this was confirmed by the report from the national materials laboratory in D. which was put in evidence. The report was to the effect that this fault, which could have arisen shortly before the bottle broke, was the direct cause of the explosion. Furthermore V. stated in his expert testimony that the explosion could be assumed to have occurred at the place of the fault. If so, the bottle had at the time of the accident a defect, which caused the bottle to explode.
If it was by reason of the chipping that the bottle exploded, then the defendant is liable for the consequent material harm to the plaintiff under 1 I (1) of the Product Liability Law. Liability could only be avoided if the defendant could prove (1 IV (2)) that at the time the bottle was put into circulation it did not have the defect, which caused the damage (1 II (2)). No such proof was adduced by the defendant in this case. It asserted only, by reference to the expert opinion of C., its production manager, that if the bottle had been chipped when it was in the factory, it would have exploded in the pressure chamber, an assertion which does not exclude the possibility that the bottle was chipped after being filled but while it was still in the defendant’s sphere of influence and risk.
On such facts there would be no room for the defence under 1 II no. 5 of the Product Liability Law: there is no difficulty in detecting such a chip.
b) But even if, as the court below evidently supposed, the bottle broke not in the area of the chip but elsewhere, by reason of a hairline crack, it was still no defence to a claim under 1 I (1) of the Law that the current state of science and technology did not permit the defect to be detected.
aa) As the court below rightly held, a product is defective under 3 I of the Law if it does not afford the safety which in all the circumstances can justifiably be expected, and consumers expect soda water bottles to be free from faults such as hairline splits and microfissures which could make them explode. The consumer’s expectation that the bottle be free from faults would not be diminished even if it were technically impossible to identify and remove such faults. The presence of such a hairline crack constitutes, as the court below rightly held, a manufacturing fault, even if it is one which “got away”. (see BGHZ 51, 91  = NJW 1969, 269 – the “chicken-pest” case).
bb) Manufacturing defects which “get away” do not, simply because they cannot be avoided by any proper precautions, constitute defects unascertainable in the current state of scientific or technical knowledge in the sense of art. 7(e) of the Directive or of 1 II no. 5 of the Law which transposes the Directive into German law.
The purpose of the rule in both instances is merely to exclude liability for what are termed development risks [references to literature]; the term covers only cases where at the time a product was put into circulation none of the means offered by the current state of science and technology rendered it possible to detect its dangerous quality. [More references]. The strict liability of the producer is to be limited by what is objectively possible in the light of the knowledge of risks available at the time the product is put into circulation. [Reference]. The only dangers to be treated as development risks are dangers inherent in the design and construction of the product, which in the current state of technology could not be avoided, not those that were inevitable at the stage of production. When the EC Directive on product liability was being fashioned it was agreed that the defence under art. 7 (e) should apply not to manufacturing defects, but only to defects of design and construction [references], and the only dangers emanating from a product which the German legislator wished to exempt from the scope of the Product Liability Law were dangers, undetectable even with the exercise of all possible care, arising at the stage of design and construction. Liability is to be excluded “only if the potential danger of the product was unrecognisable by reason of the fact that at the time of circulation it was not yet possible to recognise it” [official explanation (Gesetzeserluterung) of the draft Product Liability Law]. It is no longer a defence to this strict liability that the defective product “got away”.
The potential danger of returnable glass bottles filled with carbonated liquids has long been recognised and has indeed frequently engaged the courts [references]. As the lower court found, the danger of such glass bottles lies in the fact that even a tiny hairline crack which spreads can cause it to explode. Such a defect may arise at the stage of filling or pre-exist unnoticed, but in neither case is it a fault of design or construction, so liability in respect of it cannot be excluded under 1 II (5) of the Product Liability Law. In such a case the liability of the producer under 1 I (1) of the Law can only be avoided if it appears that the hairline crack was not in the bottle when it had been refilled and put into circulation. No such proof has been adduced by the defendant.
c) There is no need to refer the matter to the Court of Justice of the European Communities. It is true that the concept of “the state of science and technology” in 1 II no. 5 of the Product Liability Law comes from art. 7 (e) of the EC Directive and must be interpreted in a similar manner in all member states [reference]. It is also true that if the interpretation of a concept of community law is in issue the court of last resort in a member state must in principle refer the matter to the Court of Justice (art. 177 II EC). But in the present case there is no occasion to construe the concept of “the state of science”. The question is rather whether and how far the German legislator has utilised the freedom allowed by art. 15 I (b) of the Directive to deviate from art. 7(b) and make the producer liable, and this is a question for the national courts. Indeed, even where a concept in a Directive is in issue, a reference to the Court of Justice is required only if its interpretation is disputed in the literature or the case-law [reference] or if the court wishes on a point material to the case to deviate from the holding of the Court of Justice (see BVerfG NJW 1988, 2173). Neither of these is the case here.
2. The appellant is also right to criticise the court below for rejecting her claim in tort for damages for pain and suffering.
a) The court below correctly held that on the question whether a particular defect, such as the chip or hairline crack in this case, arose or even remained undiscovered while it was in the producer’s sphere of responsibility the burden of proof can be reversed if the producer was in breach of his Befundsicherungspflicht, his duty to ascertain the condition of his product and correct it if defective (BGHZ 104, 323 ; BGH NJW 1993, 528). [See note, below.] It also rightly held that users of returnable bottles are bound to have a control system which so far as is possible and reasonable in the light of the latest technology checks the condition of every bottle and takes out of circulation any bottles which might be dangerous.
b) The court below concluded that in this case the defendant had fulfilled its Befundsicherungspflicht, but there are procedural objections to the way in which it reached its conclusion.
aa) Despite the appellant’s complaint, there was no need for the court below to inquire whether the defendant excluded bottles which had been in prolonged use, since even if such a duty was broken, the breach was not causative of the harm in issue. The evidence of the national materials laboratory in D. was that the general condition of the bottle that exploded indicated that it had been used relatively infrequently. So the court below could properly suppose that the defendant was not in breach of its Befundsicherungspflicht for failing to remove from circulation, on the ground of its prior use, the bottle which injured the plaintiff.
bb) But the court below failed to obtain the further expert opinion demanded by the plaintiff on the question whether the pressure of only 5 bar in the chamber through which the bottles were passed prior to being refilled was adequate. This was a procedural lapse (references) which the plaintiff is right to criticise. The particular reason for obtaining such a report in this case is that standard spring-water bottles, which are made of thinner glass than the bottle which injured the plaintiff, are exposed to a pressure of 5.5 to 6 bar, and even this is inadequate to exclude all bottles which are apt to explode (reference). In principle the extra thickness of the bottle in question may have increased the resistance of the glass (reference), as is indeed indicated by the defendant’s assertion that 25 bar was needed to make their bottles explode. The court below should therefore have called for direct evidence on the question whether the pressure chosen by the defendant for its bottles of thicker glass was adequate to produce the desired effect of excluding damaged bottles. It was not bound to raise the pressure so high as to cause all bottles with hairline cracks to explode, for the precautions which a defendant must take in order to avoid a reversal of the burden of proof need not totally exclude the chance of explosion of bottles when handled by the consumer: it is enough if they would significantly reduce the risk of this happening (reference).
cc) The court below was also wrong to ignore the plaintiff’s evidence that the production methods used by the defendant and its system of control were inadequate to disclose existing defects such as the possible chipping of the bottle in question.
Having found that the defendant’s electronic bottle inspection unit could recognise a blemish only if it was apparent at the base or neck of the bottle, the court should not have been satisfied by the expert opinion of V. that it would be impossible to construct a machine which could detect other faults. As this court has already held (NJW 1993, 528) a mineral water company’s Befundsicherungspflicht requires it to provide a control system which reveals the condition of every single bottle and guarantees, within the limits of what is technically feasible, that dubious bottles are not reused. This does not mean that it is enough for the company to use the best possible machinery in its control procedures. If defects in bottles undetectable by machinery could be seen by human beings, the company is under a duty to arrange for a visual inspection of every single bottle. Here the defendant admittedly had the bottles inspected visually several times both before and after they were filled, but it is not clear that every bottle was so inspected, especially as during the first inspection the bottles were still in cases, and the two inspectors had to remove not only damaged bottles but also those of a different sort. Again, seeing that the throughput was 15,000 bottles per hour the subsequent inspectors had to check four bottles every second, and to check them not only for faults but also to see that they were clean, duly filled and properly labelled: it is highly unlikely that they could be expected to discover all possible faults, including a chip only a few millimetres in size. III. This being so, the decision below must be reversed. The plaintiff’s claim for material harm is ripe for final decision. The evidence shows that the explosion which damaged the plaintiff was due either to a chip or to a hairline crack, both of which are defects under 3 of the Product Liability Law so that the defendant is liable under 1 for all material harm and the declaration sought can be made. There is no case for restricting the quantum of damages since the harm suffered by the plaintiff falls below even the individual limit under 10 I of the Law.
The plaintiff’s claim for damages for pain and suffering must be reheard and decided afresh.
In the further proceedings the following must be noted. Should the court accept that the bottle was chipped, as the plaintiff claims, and the question is whether the chipping occurred during carriage or within the parents’ sphere of control and responsibility, this would be a matter for the defendant to prove if it emerged that it was in breach of its Befundsicherungspflicht. If the defendant cannot discharge that burden, it will be liable even if the explosion was not due to the chip but to a hairline crack elsewhere in the bottle: its failure to keep the bottle out of circulation would constitute a cause of the injury, and damage due to its exploding because of a hairline crack would fall within its area of responsibility since the exclusion of bottles with external damage helps protect consumers from injury through explosion. Should it transpire that the bottle was not chipped but that the explosion was due to a tiny hairline crack invisible to the human eye somewhere else on the bottle, then if the defendant has fulfilled its Befundsicherungspflicht by having adequate pressure in the pressure chamber and keeping the bottle exposed to it for long enough, it will not be liable for the plaintiff’s pain and suffering.
In product liability cases the victim must prove that the defect in the product which was the cause of the injury arose in the producer’s organisational area, before the product left the defendant’s factory. In the instant case the Bundesgerichtshof alleviates this difficult burden of proof by imposing on the defendant a Befundsicherungspflicht and applying a presumption that the defect arose in the producer’s organisational area unless he can prove that he took all possible and requisite measures of quality control to ensure that the product was free from defects. The term Befundsicherungspflichthas perplexed some commentators. Literally it signifies a duty to keep a record of the results of an investigation, such as a doctor’s note of his diagnosis, a meaning quite appropriate to its function of reversing the burden of proof in cases where the defendant can adduce relevant evidence more easily than the claimant. But the term is here used to mean a duty not just to ascertain the condition of the product but also to correct it if faulty, i.e. effectively to operate an extremely good system of quality control. As the BGH said in another case: “The producer’s duty to ascertain and assure the result of the investigation is neither a duty to “keep a record of the evidence”, as Winkelmann puts it, nor a (non-existent) duty of documentation, as Fierste suggests. In this context Befundsicherungapplies to all bottles being reused, not in the sense of making a list of the results of checking each bottle, but rather in the sense of establishing and operating a control procedure which permits the ascertainment of the physical condition of each bottle and ensures that, so far as technically possible, all bottles which are in any way faulty are kept out of further use.” (BGH 8 Dec. 1992, NJW 1993, 529.)
Finally, one should note that the device is necessary only where a claim is brought for pain and suffering under the BGB. the Directive and the laws which implement it make it clear that it is for the defendant to prove that the established defect was not present in the product when it was put into circulation.
©1998 University of Oxford. Since 2002: © Translation The University of Oxford and Professor Markesinis 1998. HTML edition © 1998 Gerhard Dannemann.